Fig 10-10

[bulli92]

in principle quite good compared to the reference figure
some deviations at very large incidence angles and for gaussian correlation function between the co-pols

[bulli92]

however, there seem to be a substantial error in the figures. When calculating kl, then one obtains 2pi/10 for l = 10cm and f=3GHz. However, the figure says it should be kl=2.1. This would correspond roughly to a frequency of 10GHz.

Thus something is wrong with the metadata in the figure!